Everything you know expires, eventually

Here’s a lesson in why experimentation and prototyping is a good idea. This is the (back end) Vasa — a 17th century Swedish warship commissioned to be the biggest and baddest thing on the high seas. And it was. Until the day the sailed it. On its maiden voyage, it floated for exactly 1500m before sinking in the Stockholm harbor. Hundreds of men died and the hull sat submerged for 333 years before being raised in the mid-20th century. If you’re in Stockholm, this is a can’t-miss museum.

Wow. Last month’s newsletter which I also published (as with all newsletters) here was extremely well received. As of this writing we’ve hit 20k reads and counting. It seems that figuring out how to reconcile the various methodologies (Agile, Lean, Design Thinking, et al) we’re embracing is something many people are working through. It also seems that you all love Goodfellas as much as I do. I’ve decided to expand on the article and self-publish a mini-book on the topic. If all goes according to plan, I should have it ready to go right after the new year.

All of these methods push us to reduce risk and validate our assumptions through short, continuous learning cycles. What’s fascinating though, is that this validated learning, if not acted upon in a timely fashion, has a half-life. In other words, it expires. Over time the validity of the learning diminishes to the point where you can’t trust the results of your experimentation and have to re-test. This is true for all knowledge it turns out. In his fascinating book, The Half-Life of Facts, Samuel Arbesman goes through, with great detail, how every truth we “know” evolves, grows, decays and dies. The same patterns that apply to scientific learning apply to the evidence we uncover with our experiments.

This is why we must act on our findings in a timely fashion. If you’re lucky enough to have continuous learning initiatives built into your product development process, you must provide your teams the autonomy to act on that insight quickly. Yes, this may disrupt your planning but it ensures that at every turn your teams are making decisions, prioritizing and shipping the best possible ideas to customers.

Here are 3 more techniques, to help reduce the decay of your learning and implement it into your delivery work:

  1. Visualize your discovery work — part of the reason validated learning decays is that it’s not treated like delivery work. To remedy this, visualize your discovery work the same way you visualize delivery work. If you write user stories and use a physical board, JIRA, Trello or any other tool, put your discovery work in the same tool. Prioritize it against the delivery work and agree, as a team, that any delivery stories that come downstream from the discovery work may have to be adjusted. This gives the discovery work the priority it needs to stay accurate and relevant.
  2. Age your discovery efforts — sometimes work takes longer to complete than planned. This is true for discovery work as well. If you’ve got a test running and it’s failing to return results in the time you planned, track its age. If it exceeds a threshold of comfort for your team, end the experiment, understand why it didn’t deliver the needed insight in time and re-run it. If a test sits too long without yielding valuable insight, the insight it is yielding is decaying steadily and will become stale.
  3. Leverage two tracks of work with one team — an anti-pattern I see in large companies is discovery work happening well upstream of delivery work. Teams call this “dual track agile” but in reality it’s just waterfall and BDUF (Big Design Up Front). If you’re going to manage a discovery track and a delivery track it has to be done with one team (the same team) and the learnings must be communicated across the team as they come in. If the feedback can’t be immediately addressed it gets shelved and must be re-validated for future iterations. Something you learned 6 months ago stands a high likelihood of not only being wrong now but potentially irrelevant.

What challenges have you had getting validated learning into your delivery funnel?



P.S. — If you’ve attended any of my workshops in the past few years you’ve likely heard my story about The Leaf Blower. It turns out, I’ve been vindicated.

— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —

Book News

The 2nd edition of Lean UX is shipping! It’s trending well in several categories on Amazon. I recommend you check it out. Sense & Respond is our follow-up to Lean UX for the leaders looking to build companies and teams that support the approaches we advocate for in that book. It is available for pre-order now. If you end up reading one or both of the books, we’d be grateful for your reviews on Amazon.

Upcoming Workshops

A few more dates in the coming months. You should join one of these workshops:

Boston, Massachusetts — Nov 16 — Lean UX Workshop at O’Reilly Training Center — O’Reilly is helping me produce a 1-day event in Boston, NEXT WEEK, focusing on building UX and design into your Agile process.

Budapest, Hungary — Stretch Leadership Conference — Dec 1–2, 2016 — I visited Budapest earlier this year for Craft conference. This event, produced by the same folks, is focused on building better teams and companies. Very excited to be a part of this.

Barcelona, Spain — Product Tank Meetup — FREE — Dec 29, 2016 — I’ll be in Barcelona on holiday late this year and thought I’d see if we could get together a group of product managers, designers and engineers to discuss the challenges we’re facing together. This event is FREE. See you there?

New York City — Feb 7–8, 2017–2-day Certified Scrum Product Owner Course with Jeff Patton. In September, we sold this course out. We’re planning another one next February back in NYC. Early bird tickets are on sale now. We’re already 20% sold out.

— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —

As always, if you want me to work directly with your company on training, coaching or workshops, don’t hesitate to reach out.